Same-sex Couples And Unmarried Partners In The American Community Survey

SAME-SEX SPOUSES AND UNMARRIED PARTNERS IN THE AMERICAN COMMUNITY SURVEY, 2008

Gary J. Gates The Williams Institute, UCLA

October 2009

Acknowledgements The author wishes to thank Lee Badgett, Brad Sears, Naomi Goldberg, and Christy Mallory for their feedback and assistance with editing.

About the Author Gary J. Gates is the Williams Distinguished Scholar at the Williams Institute, UCLA School of Law. He studies the demographic and economic characteristics of the lesbian and gay population.

SAME-SEX SPOUSES AND UNMARRIED PARTNERS IN THE AMERICAN COMMUNITY SURVEY, 2008 Executive Summary The US Census Bureau release of data from the 2008 American Community Survey (ACS) included the first official estimates for the number of same-sex couples who called one partner a “husband” or “wife”. This report compares these same-sex spousal couples to those who designated a partner as an “unmarried partner”. Comparisons are also made with comparable different-sex couples. Key findings include: 

The 2008 estimate of nearly 565,000 same-sex couples marked a decline from the peak estimate of 780,000 couples in 2006. This is likely a result of improvements made to the 2008 ACS survey instrument and in data processing procedures. o The entire decline was in the number of reported same-sex spousal couples. o The number of same-sex couples reporting themselves as unmarried couples has steadily increased from 2000 to 2008.



More than 1 in 4 (nearly 150,000) same-sex couples designated themselves as spouses. o Many same-sex couples considered themselves to be spouses even though they may not be legally married or in a legally recognized partnership. By the end of 2008, approximately 32,000 same-sex couples had been married in the US and more than 80,000 same-sex couples registered as reciprocal beneficiaries or domestic partners or were united in civil unions. These numbers fall below the estimated 150,000 same-sex spousal couples.



Same-sex spouses were identified in every state. However, they were more common in states that permit marriage for same-sex couples or offer other forms of partnership recognition. o Same-sex couples were more likely to call themselves spouses in states that recognize marriages of same-sex couples – 31% of same-sex couples in states with marriage recognition identified themselves as spouses. In states with no form of recognition, only 26% identified as spouses. o Massachusetts, the first state to permit marriage for same-sex couples in 2004, had an estimated 3.63 same-sex spousal couples per 1,000 households in 2008, ranking first among all states. Vermont, which has offered civil unions since 2000, ranked second at 2.71. The remaining top five states ranked by same-sex spouse prevalence were Hawaii (2.43), Utah (2.32), and Wyoming (2.28). o The District of Columbia had the highest prevalence of same-sex unmarried partners per 1,000 households (13.22), followed by Maine (6.81), Washington (5.84), Oregon (5.73), and New York (5.15).



Same-sex couples, both married and unmarried, were more prevalent in states that had some form of legal recognition for same-sex couples than in states that had no recognition.



Same-sex couples who identified themselves as spouses differed from same-sex couples who identified themselves as unmarried partners. o Same-sex spouses were more likely to be female; 56% of same-sex spouses were female while unmarried same-sex partners were evenly split between the sexes. This characteristic mirrors the higher rate of actual marriages by female couples in states that have extended marriage to same-sex couples. o Same-sex spouses were twice as likely to be raising children—more than 31% of spouses are raising children as opposed to 17% of unmarried partners.

i

SAME-SEX SPOUSES AND UNMARRIED PARTNERS: EXECUTIVE SUMMARY o o o

Same-sex spouses were older than unmarried partners by an average of 8.5 years. Same-sex spouses had lower education levels, employment rates, and incomes than same-sex unmarried partners. Despite their lower incomes, same-sex spouses were more likely to own their own homes than same-sex unmarried partners.



Same- sex and different-sex spouses share many characteristics. These include: o Age: the average age for same-sex spouses was 52 compared to 50 for different-sex spouses. o Education: 22% of same-sex spouses both have a college degree compared to 21% of different-sex spouses. o Income: same-sex spousal couples had an average household income of $91,558 vs. $95,075 for their different-sex counterparts. o Home ownership: 77% of same-sex spouses own their home compared to 83% of different-sex spouses. o Interracial: 7% of same-sex spousal couples are interracial compared to 6% of their different-sex counterparts



Same-sex spouses differ from different-sex spouses in child-rearing and employment rates. o 31% of same-sex spouses are raising children vs. 43% of different-sex spouses. o 46% of same-sex spousal couples have both spouses employed compared to 52% of different-sex spouses.



Same-sex unmarried partners do differ in many ways from their different-sex counterparts. They are: o Older: same-sex unmarried partners have an average age of 44 compared to 37 among different-sex unmarried partners. o More educated: same-sex unmarried partner couples are 3 times more likely to have both partners with a college education (34% vs. 10%). o Wealthier: average household incomes of same-sex unmarried partner couples are 72% higher ($112,960 vs. $65, 685) than different-sex unmarried partners. o More likely to own a home: 71% of same-sex unmarried partners own their home vs. only 45% of different-sex unmarried partners. o More likely to have both partners employed: that figure is 70% for same-sex unmarried partners vs. 62% for their different-sex counterparts. o Less likely to be raising children: 17% of same-sex unmarried partners are raising children compared to 43% of different-sex unmarried partners.



One characteristic same-sex unmarried partners share with their different-sex counterparts is similar rates of being interracial—13% of same-sex unmarried partners are interracial compared to 12% of different-sex unmarried partners.



Same-sex male and female couples were demographically similar with these notable exceptions: o More than a third (34%) of male spousal couples reported raising children compared to only 7% of male unmarried partners. Child-rearing did not vary much by spousal status for women with 28% of spouses raising children compared to 26% of unmarried partners. o While female spouses were, on average, more than four years older than male spouses; female unmarried partners were, on average, 1.4 years younger than their male counterparts. o Female spouses were less likely to be both employed (41% compared to 51% for male spouses), though employment rates were nearly the same among unmarried partners. o Female spouses and unmarried partners had lower average household income than their male counterparts. o Female couples, both spouses and unmarried partners, were less likely to be interracial.

ii

SAME-SEX SPOUSES AND UNMARRIED PARTNERS IN THE AMERICAN COMMUNITY SURVEY, 2008 Introduction The US Census Bureau release of data from the 2008 American Community Survey (ACS) provided the first official estimates for the number of same-sex couples who called one partner a “husband” or “wife”. Perhaps one of the most intriguing findings is that the estimated number of same-sex couples has declined by nearly 25% from 2007 to 2008.

term “unmarried partner” to describe one of the partners.

The 2008 ACS included several important changes that likely explain much of this decline (US Census Bureau, 2009). Both the format of the survey and various post-data processing procedures appear to have reduced respondent errors that in the past may have artificially increased the number of same-sex couples. Specifically, different-sex couples who mistakenly designated an incorrect sex for one of the partners (meaning that they appeared to be same-sex spouses) had been counted as same-sex couples.

This report includes analyses that offer evidence that improved procedures used in the 2008 ACS have minimized the possibility that a large portion of same-sex spouses are actually miscoded different-sex married couples. While these data offer the best opportunity to date to consider differences between same-sex spousal couples and those who consider themselves to be unmarried partners, it is still possible that some miscoded different-sex couples may be included among the same-sex spouses analyzed in this report.

Analyses of data from Census 2000 and ACS data from 2005 to 2007 (Black et al., 2007; Gates and Steinberger, 2009) suggest that as many as three-quarters of same-sex couples who used the term “husband/wife” may have actually been miscoded different-sex married couples. Findings described in this report provide evidence that improvements in survey design and processing may have substantially reduced this problem.

The report begins by showing how estimates for the number of same-sex spouses and unmarried partners have changed over time and considers how improvements in the 2008 ACS may help to explain these changes. That is followed by an analysis of the geographic distribution of samesex spouses and unmarried partners, including consideration of how state laws regarding relationship recognition might affect the presence of same-sex couples. Demographic analyses follow and consider differences among same-sex and different-sex spousal couples and unmarried partners and differences between male and female same-sex couples.

It is important to note that this does not mean that prior Williams Institute demographic studies using these data, particularly the “Census Snapshot” series, are inherently flawed (e.g., Romero et al., 2007; Gates and Ramos, 2008). While estimates of the number of samesex couples from prior studies may be now viewed as too high, analyses of demographic characteristics in all Williams Institute studies adjusted data so as to effectively remove samesex spouses (thus removing any miscoded different-sex spouses). Analyses in these prior studies include only those couples who used the

Readers seeking to compare demographic traits from past Williams Institute studies with those in these analyses should compare traits of only those couples who used the “unmarried partner” designation.

1

SAME-SEX SPOUSES AND UNMARRIED PARTNERS married couples who made a mistake and checked an incorrect sex box for one of the spouses (which resulted in being mistakenly counted among same-sex couples). The presence of miscoded different-sex married couples artificially inflated the number of samesex couples.

By the numbers Changes over time The 2008 American Community Survey identified an estimated 564,743 same-sex couples in the United States, including 149,956 couples who designated themselves as spouses and 414,787 who used the term unmarried i partner (see Figure 1). A detailed explanation of the data used in this report, including how same-sex couples are identified, is included in the Appendix.

The 2008 ACS included a major and improved redesign of both the survey and post-data collection editing techniques that will conform to standards that also will be used in the upcoming Census 2010 (US Census Bureau, 2009). These changes likely reduced the rate of sex miscoding among different-sex spouses. The entire reduction in the estimated number of same-sex couples was a result of large declines among same-sex spouses, presumably because they are comprised of fewer different-sex married couples who miscoded the sex of one spouse.

The total number of same-sex couples has declined from 2005-2007 ACS estimates that have exceeded 750,000. This is likely a result of changes in the format and processing of the 2008 ACS that reduced the probability of ii respondents making errors. Gates and Steinberger (2009) analyzed recent ACS data and demonstrated that a very large fraction of same-sex couples who likely referred to themselves as spouses were not actually same-sex. Instead, they were different-sex

Figure 1 Same-sex spouses and unmarried partners Census 2000 and American Community Survey 2005-2008 776,943

779,867

753,618

594,391 564,743

384,629392,314

394,115385,752

341,014

412,770

414,787

340,848

253,377

149,956

Census 2000

ACS 2005

ACS 2006

Unmarried Partners

2

ACS 2007 Spouses

Total

ACS 2008

SAME-SEX SPOUSES AND UNMARRIED PARTNERS Same-sex spouses and legal recognition of same-sex couples The estimate of 150,000 same-sex spousal couples far exceeds the number of legally married same-sex couples in the United States, which is approximately 32,000 (see Appendix iii Table 2) . The figure also exceeds the number of same-sex couples who are in other forms of legal recognition like civil unions and registered domestic partnership (see the Appendix for detailed information about legal recognition for same-sex couples). Figure 2 Same-sex couples designating one partner as a husband/wife by state legal recognition for same-sex couples

31.3% 28.8% 25.6%

States with no recognition

States with marriage or other legal recognition

States with marriage

Clearly the designation of a spouse within a same-sex couple is not necessarily an indication that the couple has been legally married. It may be that some same-sex couples have been united in religious ceremonies or commitment ceremonies and consider themselves to be married, regardless of legal status. It may also be that some same-sex couples simply believe that terms like husband and wife provide the best description of their relationship regardless of any form of legal recognition. Nevertheless, evidence exists to show that same-sex couples who live in states that provide marriage equality or other forms of legal recognition are more likely to use spousal terms to describe themselves. The proportion of same-sex couples who designate one partner as a spouse is highest (31.3%) in states that offer legal marriage to those couples (see Figure 2). In states with any form of legal recognition (marriage, civil union, registered domestic

partnership, or reciprocal beneficiary), 28.8% of same-sex couples included a spouse and in states with no recognition the figure was only 25.6%. Somewhat more striking is that the prevalence of same-sex spousal couples per 1,000 households (see Figure 3) in states with marriage equality (2.17) and those with some form of legal recognition (1.95) is nearly twice as large as the prevalence in states with no legal recognition (1.16). The sex composition of same-sex spouses also offers evidence that the 2008 figures captured a larger portion of couples who are in legally recognized relationships. Women comprise nearly two-thirds of same-sex couples who seek legal relationship recognition in the US (Gates et al., 2008). Prior to the 2008 ACS, Census and ACS data have shown that same-sex unmarried partners were generally split evenly between male and female couples while spouses tended to be majority male (see Figure 4). In the 2005 through 2007 ACS estimates, women comprised between 44 and 45% of same-sex spouses. That pattern changed in 2008 ACS estimates. While unmarried partners were still split evenly between men and women, women comprised 56% of same-sex spouses. This puts these estimates more in line with the sex composition of same-sex married couples and those who have sought other forms of legal recognition. Figure 3 Same-sex spouses per 1,000 households by state legal recognition for same-sex couples

2.17 1.95

1.16

States with no recognition

States with marriage or other legal recognition

States with marriage

3

SAME-SEX SPOUSES AND UNMARRIED PARTNERS

Geographic distribution

Figure 4 % Female among Same-sex Spouses

State rankings

Census and American Community Survey

Further evidence of a link between legal recognition and spousal designation can be seen in state rankings by prevalence of same-sex spousal couples per 1,000 households. All three states offering marriage to same-sex couples and three other states with non-marital legal recognition ranked in the top ten. Not surprisingly, Massachusetts, which has permitted same-sex couples to marry since 2004, ranked first with 3.63 same-sex spousal couples per 1,000 households. California and Connecticut, the two other states offering legal marriage to same-sex couples in 2008, were th th ranked 6 and 8 , respectively. Vermont, which has permitted civil unions since 2000, nd ranked 2 and Hawaii and New Jersey, which also provide legal recognition through reciprocal beneficiary status and civil unions, respectively, rd th also ranked in the top ten at 3 and 9 , respectively. Four states with no legal recognition—Utah, Wyoming, Nevada, and Rhode Island—also ranked in the top ten. Among same-sex couples who used the unmarried partner designation, California was the only state offering marriage that ranked th among the top ten (ranking 6 ) in prevalence per 1,000 households. The top five states—DC, Maine, Washington, Oregon, and New York—all offered some form of legal recognition for same-sex couples.

48.7%

50.2%

45.3%

Census 2000

ACS 2005

50.6%

50.2%

49.4%

48.4%

56.1%

44.3%

43.6%

ACS 2006 Unmarried Partners

ACS 2007 Spouses

When same-sex spouses and unmarried partners are combined and ranked by prevalence per 1,000 households, eight of the top ten states had some form of legal relationship recognition for same-sex couples in 2008. Of the two states that did not have recognition, Colorado enacted “designated beneficiary” legislation in 2009 providing some legal protections for same-sex couples.

Table 1. Same-sex couples per 1,000 households, by spouses and unmarried partner designation

Rank 1 2 3 4 5 6 7 8 9 10

Same-sex spouses per 1,000 households Massachusetts 3.63 Vermont 2.71 Hawaii 2.43 Utah 2.32 Wyoming 2.28 California 1.92 Nevada 1.85 Connecticut 1.79 New Jersey 1.70 Rhode Island 1.64

Same-sex unmarried partners per 1,000 households District of Columbia 13.22 Maine 6.81 Washington 5.84 Oregon 5.73 New York 5.15 California 5.01 Colorado 4.92 Arizona 4.65 Delaware 4.59 Rhode Island 4.41

State offered marriage to same-sex couples in 2008 State offered non-marital relationship recognition for same-sex couples in 2008 State recognized same-sex marriages in 2008 but they could not be performed there

4

Same-sex couples (spouses and unmarried partners) per 1,000 households District of Columbia 14.12 Maine 8.23 Massachusetts 7.92 Oregon 7.26 Washington 6.97 California 6.93 New York 6.41 Colorado 6.13 Vermont 6.10 Delaware 6.09

ACS 2008

SAME-SEX SPOUSES AND UNMARRIED PARTNERS Figure 5

Figure 6

5

SAME-SEX SPOUSES AND UNMARRIED PARTNERS National distribution of same-sex spouses and unmarried partners Same-sex spouses were identified in every state. However, the proportion of same-sex spouses among all households varied considerably across states (see Figure 5). Same-sex spouses were most prevalent in the Northeast, especially New England, and the West and Mountain states. Same-sex spouses were generally least prevalent in Midwest and upper Mountain states. Similar to their spousal counterparts, same-sex unmarried partners were most prevalent in the Northeast and along the West coast (see Figure 6). They were least prevalent in the upper Mountain and Midwest states and in the Deep South. Prevalence figures for all states are shown Appendix Table 2.

Demographic characteristics Differences by spouse/unmarried partner status and sex composition For many demographic characteristics, the patterns observed in differences between spousal couples and unmarried partners were similar for both different-sex and same-sex couples. Compared to unmarried partners, those in spousal couples (both same-sex and different-sex) were older, less likely to be interracial, less likely to be both employed, and more likely to own their homes. There were some characteristics where the patterns differed by the sex composition of the couples. For example, child-rearing did not differ between different-sex spousal couples and unmarried partners. However, same-sex spousal couples were more likely to be raising children than same-sex unmarried partners. Different-sex spousal couples had higher levels of education than their unmarried partner counterparts. The reverse was true for samesex couples where spouses had lower education levels than unmarried partners. Consistent with this finding, different-sex spouses had higher incomes than different-sex unmarried partners while same-sex spouses had lower incomes than their unmarried partner counterparts.

6

In some cases, there were differences between same-sex and different-sex couples, regardless of spousal status. Same-sex couples had higher education levels and were less likely to be raising children than were different-sex couples. Appendix Table 1 includes all demographic data analyzed in this report.

Age When compared to their different-sex counterparts, same-sex spouses and unmarried partners were, on average, older. It should be noted that it is not true that those in same-sex couples in general are older than those in different-sex couples. When same-sex couples are combined to include same-sex spouses and partners, they are generally younger than different-sex spouses and older than differentsex unmarried partners because most same-sex couples report themselves as partners. Age differences between same-sex and different-sex couples were particularly large when comparing unmarried partner couples (see Figure 7). Among unmarried partners, those in same-sex couples were, on average, nearly seven years older than different-sex householders and partners. Among spousal couples, those in same-sex couples were, on average, nearly three years older than their different-sex counterparts.

Figure 7 Average age by sex composition and spousal status of couples 52.4 49.7 43.9 37.0

Different-sex

Same-sex Spousal couples

Different-sex

Same-sex

Unmarried partner couples

SAME-SEX SPOUSES AND UNMARRIED PARTNERS Like different-sex couples, those who identified as spouses in same-sex couples were also older than those who identified as unmarried partners. However, the difference in average age between same-sex spouses and unmarried partners (8.5 years) was not as large as the difference between different-sex spouses and unmarried partners (12.7 years).

Figure 9 % Both partners are employed by sex composition and spousal status of couples

70% 62% 52% 46%

Education, Employment and Income Compared to different-sex spouses, same-sex spouses were only slightly more likely to both have a college degree, 21% vs. 22%, respectively (see Figure 8). However, same-sex unmarried partners were substantially more likely to both have a college degree than their different-sex counterparts, 34% vs. 10%, respectively. Figure 8 % Both partners have at least a college degree by sex composition and spousal status of couples

34%

22%

21%

10%

Different-sex

Same-sex Spousal couples

Different-sex

Same-sex

Unmarried partner couples

Employment patterns differed between spousal and unmarried partner couples. Among spousal couples, different-sex couples were more likely to be both currently employed than their samesex counterparts, 52% vs. 46%, respectively (see Figure 9). The reverse was true for unmarried partner couples where same-sex couples were more likely than different-sex couples to be both employed, 70% vs. 62%, respectively. The patterns observed regarding age, education, and employment all help to explain differences in household income levels (see Figure 10). Same-sex spouses were close in both age and education to their different sex counterparts, though they were somewhat less likely to be employed. This lower level of employment may in part explain why their average household

Different-sex

Same-sex Spousal couples

Different-sex

Same-sex

Unmarried partner couples

income was approximately $3,500 lower than that of different-sex spouses. It may also be related to the fact that women comprise 56% of the same-sex spouses. Female same-sex spousal couples had an average annual household income of only $86,000 while their male counterparts had an average household income of nearly $99,000 (see Appendix Table 2). Among unmarried partner couples, those in same-sex couples were older, substantially more educated, and had higher levels of employment than their different-sex counterparts. These factors likely all contributed to a sizable difference in average household income. Samesex unmarried partner couples reported average annual household income of nearly $112,960 compared to only $65,685 among different-sex unmarried partner couples. Figure 10 Average household income by sex composition and spousal status of couples

$112,960 $95,075

$91,558

$65,685

Different-sex

Same-sex Spousal couples

Different-sex

Same-sex

Unmarried partner couples

7

SAME-SEX SPOUSES AND UNMARRIED PARTNERS Figure 11 % Own their home

Child-rearing

by sex composition and spousal status of couples

83% 77% 71%

45%

Same-sex couples were less likely to be raising children than their different-sex counterparts (see Figure 13). However, same-sex spouses were more likely to be raising children than same-sex unmarried partner couples. While 43% of different-sex couples (spousal and unmarried partner) were raising children, 31% of same-sex spouses and 17% of same-sex unmarried partners were doing the same. Figure 13 % Raising children

Different-sex

Same-sex

Different-sex

Spousal couples

by sex composition and spousal status of couples

Same-sex

Unmarried partner couples

43%

43%

Home ownership Consistent with the pattern observed with differences in household incomes, same-sex spouses were slightly less likely than differentsex spouses to own their homes, 83% vs. 77%, respectively (see Figure 11). Also consistent with income patterns, same-sex unmarried partners were substantially more likely to own their homes than their different-sex counterparts, 71% vs. 45%, respectively.

Same-sex spouses and unmarried partners were more likely to be interracial than their different sex counterparts (see Figure 12). Among spousal couples, 7% of same-sex couples were interracial compared to less than 6% of different-sex couples. Among unmarried partners, 12.7% of same-sex couples were interracial compared to 12% of different-sex couples. Figure 12 % Interracial by sex composition and spousal status of couples

12.7% 12.0%

7.1% 5.9%

Same-sex Spousal couples

8

17%

Different-sex

Same-sex Spousal couples

Racial composition

Different-sex

31%

Different-sex

Same-sex

Unmarried partner couples

Different-sex

Same-sex

Unmarried partner couples

Differences between male and female same-sex couples Across most demographic characteristics, patterns observed between spouses and unmarried partners were relatively similar for male and female couples. For both male and female couples, spouses were older, were less likely to be interracial, had lower levels of education, were less likely to be employed, were more likely to own their home, had lower household incomes, and were more likely to have children than their unmarried partner counterparts (see Figure 14). There were a few notable differences between same-sex male and female couples. While female spouses were, on average, more than four years older than male spouses, female unmarried partners were, on average, 1.4 years younger than their male counterparts. Female couples, both spouses and unmarried partners, were less likely to be interracial. Female spouses were less likely to be both employed

SAME-SEX SPOUSES AND UNMARRIED PARTNERS (41% compared to 51% for male spouses) though employment rates were nearly the same among unmarried partners.

Male spouses actually reported higher levels of child-rearing than their female counterparts. But child-rearing differences between spouses and unmarried partners were much greater for male couples than female couples. More than a third (34%) of male spousal couples reported raising children compared to only 7% of male unmarried partners. Child-rearing did not vary much by spousal status for women with 28% of spouses raising children compared to 26% of unmarried partners.

Despite having relatively similar age, education, and employment levels, female unmarried partners had substantially lower average household income than their male counterparts. Female spouses also had lower average household income than their male counterparts. These differences by sex may in part be explained by lower levels of employment among female spouses and by the gender wage gap.

Figure 14 Selected demographic characteristcs by sex and spousal status of same-sex couples $129,607

$98,730

$96,719

$85,956 79% 75%

72% 71%

70% 69%

54.3

51%

50.0 44.6 43.2

41% 34%

34% 34%

28%

26%

22% 21% 14% 8%

Average age

11%

7%

6%

Interracial

Male Spouses

Both partners with a college degree Female Spouses

Both partners employed

Own home

Male unmarried partners

Average Household Income (dollars)

Raising children

Female unmarried partners

9

SAME-SEX SPOUSES AND UNMARRIED PARTNERS

Appendix The ACS and enumeration of same-sex couples This report analyzes US Census Bureau data regarding same-sex spouses and unmarried partners from the 2008 American Community Survey (ACS). This marks the first time that the Census Bureau has included separate estimates of same-sex spouses in their official data releases from either decennial Census data iv or the ACS. Conducted every ten years, the US Census collects data from all US households. The ACS is an annual survey that bases its population estimates on a sample of approximately two million households. In both the Census and the ACS, same-sex couples are identified in households where the “householder” (the person who rents or owns the home) identifies another adult of the same sex as either a “husband/wife” or an “unmarried partner”. The “unmarried partner” category was first introduced in the 1990 decennial census (Black et al. 2000). In general, this term was designed to capture couples (both same-sex and different-sex) who are in a “close personal relationship” and are not legally married. This marked the first opportunity to count same-sex cohabiting couples in these close relationships. Same-sex spouses have been enumerated in different ways since 1990. The Census Bureau routinely edits data to correct obvious errors and create consistency in data reporting. In 1990, the Census Bureau assumed that same-sex spouses were most likely different-sex married couples who made a mistake and incorrectly coded the sex of one of the spouses. So they edited the sex of the same-sex spouse and changed the household to a different-sex married couple. In Census 2000 and in the ACS prior to 2008 the decision was made to no longer change the sex of the spouse but instead designate the same-sex spouse as a same-sex unmarried partner. So counts of same-sex unmarried partners included same-sex couples who designated one partner as either a spouse or an unmarried partner. Unfortunately, this meant that these counts also include different-sex married couples who miscoded the sex of one of the spouses. There have been a variety of publications using data from Census 2000 and the ACS (e.g., Gates and Ost, 2004; Romero et al. 2007; Gates and Ramos, 2008). In all cases of Williams Institute studies, demographic data were adjusted to try to eliminate miscoded different-sex married couples from the data. As such, the demographic characteristics reported in these past reports are best compared only to the characteristics of same-sex couples who explicitly used the “unmarried partner” designation in this report using the 2008 ACS data. The 2008 ACS data release included separate estimates for the number of same-sex spouses and unmarried partners along with selected demographic characteristics of these two groups.

Legal recognition of same-sex couples Hawaii was the first state in the United States to recognize same-sex couples in the form of reciprocal beneficiary relationships in 1997. Since then, 14 other states and the District of Columbia have v established same-sex legal relationships in the form of marriage, civil unions, and domestic partnerships. As shown in Appendix Table 2, by the end of 2008, an estimated 31,829 same-sex couples had been legally married and 87,380 same-sex couples had formalized their relationships in another legal fashion in the United States (domestic partnership, civil union, or reciprocal beneficiary). Some couples who registered as domestic partners may have subsequently obtained married licenses, so it is not possible to add these figures together to determine how many same-sex couples are have a legally recognized relationship. Additionally, in D.C., Hawaii, and Maine, California, New Jersey, and

SAME-SEX SPOUSES AND UNMARRIED PARTNERS Washington, both different-sex and same-sex couples are permitted to register as reciprocal beneficiaries vi or domestic partners. Appendix Table 1. Demographic characterstics by couple type, 2008 American Community Survey. Married Different-sex Couples

Unmarried Different-sex Couples

Same-sex Couples (spouses and unmarried partners)

55,692,136

Total

5,648,999

564,743

Average Age Both Partners have at least a college degree Both Partners are employed

49.7

37.0

46.2

21.1%

9.8%

30.6%

51.6%

61.6%

63.5%

Average Household Income

$95,075

$65,685

$107,277

Own home

82.5%

45.2%

72.8%

Interracial Couple

5.9%

12.0%

11.2%

43.2%

43.1%

20.5%

Same-Sex Spouses

Same-Sex Unmarried Partners

Raising children

Total

All 149,956

Male 65,764

Female 84,192

All 414,787

Male 204,836

Female 209,951

Average Age Both Partners have at least a college degree Both Partners are employed

52.4

50.0

54.3

43.9

44.6

43.2

21.7%

22.0%

21.5%

33.8%

33.7%

33.9%

45.5%

51.3%

41.0%

69.9%

70.4%

69.5%

Average Household Income

$91,558

$98,730

$85,956

$112,960

$129,607

$96,719

Own home

77.2%

74.6%

79.2%

71.3%

71.7%

70.8%

Interracial Couple

7.1%

8.5%

6.0%

12.7%

14.1%

11.4%

Raising children

30.5%

33.9%

27.9%

16.8%

7.4%

25.9%

SAME-SEX SPOUSES AND UNMARRIED PARTNERS

Appendix Table2. Same-sex spouses, unmarried partners, and legally recognized marriages and other forms of legal recognition, by state. Same-sex couples

United States Alabama

Total 564,743

Per 1,000 households 4.99

Same-sex spouses

47

Total 149,956

Per 1,000 households 1.33

Rank

Same-sex unmarried partners

35

Total 414,787

Per 1,000 households 3.67

Rank

Rank 47

4,850 854

2.67

41

1,856

1.02

25

2,994

1.65

3.59

12

277

1.17

33

577

2.43

8

12,960

5.70

46

2,384

1.05

34

10,576

4.65

45

Arkansas

3,176

2.85

6

1,158

1.04

6

2,018

1.81

6

California

84,397

6.93

8

23,403

1.92

23

60,994

5.01

7

Colorado

Alaska Arizona

40

11,635

6.13

17

2,304

1.21

8

9,331

4.92

18

Connecticut

6,865

5.16

10

2,377

1.79

13

4,488

3.38

9

Delaware

2,003

6.09

1

494

1.50

40

1,509

4.59

1

District of Columbia

Through 2008 Legal Non-marital marriages legal (same-sex) recognition 31,829 87,380

3,529

14.12

14

225

0.90

19

3,304

13.22

12

Florida

39,641

5.62

16

9,149

1.30

22

30,492

4.32

14

Georgia

18,181

5.24

13

4,218

1.22

3

13,963

4.02

22

Hawaii

2,472

5.66

42

1,060

2.43

43

1,412

3.23

41

Idaho

1,840

3.25

20

484

0.86

17

1,356

2.40

20

Illinois

22,141

4.65

32

6,412

1.35

29

15,729

3.30

31

Indiana

10,058

4.05

34

2,675

1.08

15

7,383

2.98

38

Iowa

4,817

3.96

44

1,716

1.41

51

3,101

2.55

39

Kansas

3,348

3.01

36

645

0.58

27

2,703

2.43

34

Kentucky

6,581

3.90

43

1,926

1.14

31

4,655

2.76

44

Louisiana

5,143

3.16

2

1,742

1.07

14

3,401

2.09

2

Maine

4,461

8.23

22

769

1.42

44

3,692

6.81

15

Maryland

9,361

4.47

3

1,788

0.85

1

7,573

3.62

13

Massachusetts

19,550

7.92

40

8,951

3.63

32

10,599

4.30

37

Michigan

13,774

3.61

35

4,042

1.06

46

9,732

2.55

29

Minnesota

8,218

3.93

51

1,731

0.83

47

6,487

3.10

50

Mississippi

2,360

2.16

33

900

0.82

45

1,460

1.33

26

Missouri

9,384

4.03

39

1,944

0.83

41

7,440

3.19

35

Montana

1,366

3.64

45

338

0.90

49

1,028

2.74

42

18,000

51,497

544

2,099 802*

1,570

982 13,285

SAME-SEX SPOUSES AND UNMARRIED PARTNERS

Nebraska

2,087

2.96

18

490

0.70

7

1,597

2.27

23

Nevada

4,820

5.06

24

1,760

1.85

48

3,060

3.21

16

New Hampshire

2,192

4.34

19

388

0.77

9

1,804

3.57

24

621

New Jersey

15,443

4.90

15

5,368

1.70

21

10,075

3.19

11

14267

New Mexico

4,157

5.61

7

911

1.23

20

3,246

4.38

5

New York

45,761

6.41

27

8,981

1.26

26

36,780

5.15

30

North Carolina

15,315

4.26

50

4,155

1.16

50

11,160

3.10

46

649

2.36

29

178

0.65

38

471

1.71

27

18,854

4.18

37

4,512

1.00

42

14,342

3.18

33

5,189

3.69

4

1,264

0.90

11

3,925

2.79

4

Oregon

10,704

7.26

28

2,256

1.53

18

8,448

5.73

32

Pennsylvania

20,656

4.21

11

6,487

1.32

10

14,169

2.89

10

Rhode Island

2,414

6.05

30

655

1.64

12

1,759

4.41

36

South Carolina

7,050

4.14

49

2,579

1.52

16

4,471

2.63

51

777

2.43

25

448

1.40

37

329

1.03

19

Tennessee

10,546

4.33

23

2,478

1.02

24

8,068

3.31

21

Texas

37,557

4.46

21

10,065

1.20

4

27,492

3.26

43

Utah

3,861

4.52

9

1,984

2.32

2

1,877

2.20

17

Vermont

1,524

6.10

26

678

2.71

30

846

3.38

25

Virginia

12,639

4.27

5

3,184

1.08

28

9,455

3.19

3

Washington

17,756

6.97

48

2,873

1.13

36

14,883

5.84

48

West Virginia

1,902

2.54

31

764

1.02

39

1,138

1.52

28

Wisconsin

9,166

4.07

38

2,053

0.91

5

7,113

3.16

49

Wyoming

759

3.64

47

476

2.28

35

283

1.36

47

North Dakota Ohio Oklahoma

South Dakota

*data only through April 2008

2,636

8,903 4,003

SAME-SEX SPOUSES AND UNMARRIED PARTNERS

References Black, D, Gates, GJ, Sander, S, Taylor, L. Demographics of the Gay and Lesbian Population in the United States: Evidence From Available Systematic Data Sources, Demography, May 2000, 37(2) 139-54. Black, D, Gates, GJ, Sander, S, Taylor, L. The Measurement of Same-Sex Unmarried Partner Couples in the 2000 U.S. Census, California Center for Population Research (working paper series, UCLA, September 2007. http://papers.ccpr.ucla.edu/papers/PWP-CCPR-2007-023/PWP-CCPR-2007-023.pdf Gates, GJ and Ost, J. The Gay & Lesbian Atlas. Urban Institute Press: Washington DC, 2004. Gates, GJ. Geographic Trends Among Same-Sex Couples in the U.S. Census and the American Community Survey, Williams Institute, UCLA School of Law, November 2007 (http://www.law.ucla.edu/williamsinstitute/publications/ACSBriefFinal.pdf). Gates, GJ, Badgett, MVL, Ho, D. Marriage, Registration and Dissolution by Same-sex Couples in the U.S., Williams Institute, UCLA School of Law, July 2008 (http://www.law.ucla.edu/williamsinstitute/publications/Couples%20Marr%20Regis%20 Diss.pdf). Gates, GJ, Ramos, C. Census Snapshot: California Lesbian, Gay, and Bisexual Population, Williams Institute, UCLA School of Law, October 2008. http://www.law.ucla.edu/williamsinstitute/publications/CA%20Snapshot%202008.pdf Gates, GJ and Steinberger, M. “Same-Sex Unmarried Partner Couples in the American Community Survey: The Role of Misreporting, Miscoding and Misallocation,” Paper presented at the Annual Meetings of the Population Association of America, Detroit, MI, April 30, 2009. O’Connell, M and Lofquist, D. “Counting Same-Sex Couples: Official Estimates and Unofficial Guesses,” Paper presented at the Annual Meetings of the Population Association of America, Detroit, MI, April 30, 2009. Romero, AP, Baumle, AK, Badgett, MVL, Gates, GJ. Census Snapshot: United States, Williams Institute, UCLA School of Law, December 2007. http://www.law.ucla.edu/williamsinstitute/publications/USCensusSnapshot.pdf US Census Bureau. Changes to the American Community Survey between 2007 and 2008 and their Potential Effect on the Estimates of Same-Sex Couple Households, Families and Living Arrangements, 2009. (http://www.census.gov/population/www/socdemo/files/changes-to-acs-2007-to2008.pdf) i

Separate figures for same-sex unmarried partners and spouses in Census 2000 and the ACS 2005-2007 were reported in O’Connell and Ludquist (2009). The remainder of the data for this report comes from this table: http://www.census.gov/population/www/socdemo/files/ssex-tables-2008.xls.

SAME-SEX SPOUSES AND UNMARRIED PARTNERS

ii

For additional information about changes in the 2008 American Community Survey, see “Changes to the American Community Survey between 2007 and 2008 and their Potential Effect on the Estimates of Same-Sex Couple Households” (http://www.census.gov/population/www/socdemo/files/changes-to-acs-2007-to-2008.pdf). iii

Estimates for the number of legal same-sex marriages are derived from administrative data reported from all states that permit marriage for same-sex couples (through 2008) except California, where marriage license forms did not designate the sex of the spouses. The estimate for the number of same-sex couples married in California is based on Williams Institute analyses of all marriages conducted in similar time periods in 2007 and 2008. Increases in the total number of marriages provide the basis for estimating the number of same-sex couples who married. iv

Estimated counts for the number of same-sex spouses in Census 2000 and in the ACS 2005-2007 were reported in O’Connell and Lofquist (2009). They were not available when those data were originally released. v

Marriage: Massachusetts (2004), Connecticut (2008), California (June 2008 – November 2008) Iowa (2009), Maine (2009), Vermont (2009), and New Hampshire (2010); Domestic Partnerships (state-wide rights equivalent to marriage): California (1999, expanded 2005), District of Columbia (2005), Oregon (2008), Washington (2007, expanded 2009), and Nevada (2009); Civil Union (state-wide rights equivalent to marriage): New Jersey (2007); Limited recognition: Hawaii (1997), Colorado (2009), and Wisconsin (2009). vi

In personal communication with UCLA professor Susan Cochran, she found that 95% of registrants in California were likely samesex couples.

For more information, contact: The Williams Institute UCLA School of Law Box 951476 Los Angeles, CA 90095-1476 T (310) 267-4382 F (310) 825-7270 [email protected] www.law.ucla.edu/williamsinstitute